OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Vectorizer

Vectorizer is a Rust-based, self-hosted in-memory vector database that stores embeddings and supports semantic similarity and top-k nearest-neighbor queries, with binary file persistence.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Vectorizer — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Vectorizer when an application needs a self-hosted index for embedding storage and similarity retrieval, and its documented transport and deployment options fit the surrounding system.

Who it is for

  • Developers adding semantic search to an application
  • Teams building recommendation, matching, or embedding-backed knowledge retrieval
  • Operators who want to run a vector database on their own infrastructure

Documented uses

  • Store application-generated embeddings in a searchable index
  • Retrieve top-k nearest vectors for matching or recommendations
  • Search using cosine, Euclidean, or dot-product similarity
  • Persist vector data in binary files
Before you choose

Requirements & limits

The repository is public and identifies Vectorizer as Apache-2.0 licensed. The README documents installation scripts for Linux, macOS, and Windows, as well as Docker deployment.

  • The repository describes Vectorizer as an in-memory database, while also documenting binary persistence and memory-mapped storage.
  • Connection choices include binary MessagePack over TCP, REST, gRPC, GraphQL, and MCP.
  • The default Docker image is described as BM25-only; the -fastembed image is documented for dense and multilingual models.
  • Docker authentication is enabled by default, and the README specifies minimum production lengths for the JWT secret and administrator password.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Vectorizer with Docker

Start the documented Docker service with a persistent data volume and authentication configured.

Before you start

  • Docker
  • A secure administrator password of at least 32 characters
  • A JWT secret of at least 32 characters; the README shows generating one with openssl

Steps

  1. Run the README's Docker command using the ghcr.io/hivellm/vectorizer:3.8.2 image.

  2. Publish ports 15002 and 15503 and mount a named data volume at /data.

  3. Set VECTORIZER_AUTH_ENABLED=true, VECTORIZER_ADMIN_USERNAME, VECTORIZER_ADMIN_PASSWORD, and VECTORIZER_JWT_SECRET.

Expected outcome

A Vectorizer container is configured with a persistent data volume and authentication enabled.

Guide limitations

  • The command documents service startup, not subsequent collection creation or search requests.
  • The default image is described as BM25-only; the -fastembed image is documented for dense and multilingual models.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for Vector search

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
VectorizerStore application-generated embeddings in a searchable indexSelf-hosted, ServerThe repository is public and identifies Vectorizer as Apache-2.0 licensed. The README documents installation scripts for Linux, macOS, and Windows, as well as Docker deployment.
MettisIndex text documents through an HTTP endpoint.Linux, Self-hosted, Server
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
hivellm/vectorizer · Official homepage title · retained in its source language
How we review sourcesSuggest a correction